Glossary · term

GenAI Agent Spans

An extension of OpenTelemetry's semantic conventions for generative applications, describing agent and framework operations as distributed-system traces (spans). It defines attributes and operation names such as create_agent, invoke_agent, and execute_tool, so that observability covers the agent's entire operation rather than just individual model calls.

LLMOps2026Wave 3 · 2025–26Maturity: 2/5

Maturity rationale

single source, early stage

References

Oficjalna specyfikacja OTel; szerokie wsparcie wendorskie (Datadog, repo open-te(spec)

Author: METR

Architecture & engineering guide

Read the in-depth implementation guide: AgentSpec and OpenTelemetry GenAI: Runtime Enforcement and Observability for AI Agents.

Cite or link to this resource · CC BY 4.0

Free to quote, link, and reuse with attribution

Referencing this analysis in documentation, research, or an engineering write-up? Copy a formatted link or citation below:

Show citation text & canonical URL

Markdown:

[GenAI Agent Spans — AI Glossary — Skills Intelligence](https://www.skillsintelligence.tools/glossary/term/genai-agent-spans)

APA:

Skills Intelligence. (2026). GenAI Agent Spans — AI Glossary. Skills Intelligence AI Glossary. Retrieved from https://www.skillsintelligence.tools/glossary/term/genai-agent-spans